There is a problem with the pre-training weights

[softalter]

scripts/train.py --dataset_type cars_encode --exp_dir directory --use_w_pool --w_discriminator_lambda 0.1 --progressive_start 20000 --id_lambda 0.5 --val_interval 10000 --start_from_latent_avg --max_steps 200000 --stylegan_size 512 --stylegan_weights a/stylegan2-ffhq-config-f.pt --workers 8 --batch_size 8 --test_batch_size 4 --test_workers 4
{'batch_size': 8,
'board_interval': 50,
'checkpoint_path': None,
'd_reg_every': 16,
'dataset_type': 'cars_encode',
'delta_norm': 2,
'delta_norm_lambda': 0.0002,
'encoder_type': 'Encoder4Editing',
'exp_dir': 'directory',
'id_lambda': 0.5,
'image_interval': 100,
'keep_optimizer': False,
'l2_lambda': 1.0,
'learning_rate': 0.0001,
'lpips_lambda': 0.8,
'lpips_type': 'alex',
'max_steps': 200000,
'optim_name': 'ranger',
'progressive_start': 20000,
'progressive_step_every': 2000,
'progressive_steps': [0,
20000,
22000,
24000,
26000,
28000,
30000,
32000,
34000,
36000,
38000,
40000,
42000,
44000,
46000,
48000],
'r1': 10,
'resume_training_from_ckpt': None,
'save_interval': None,
'save_training_data': False,
'start_from_latent_avg': True,
'stylegan_size': 512,
'stylegan_weights': 'a/stylegan2-ffhq-config-f.pt',
'sub_exp_dir': None,
'test_batch_size': 4,
'test_workers': 4,
'train_decoder': False,
'update_param_list': None,
'use_w_pool': True,
'val_interval': 10000,
'w_discriminator_lambda': 0.1,
'w_discriminator_lr': 2e-05,
'w_pool_size': 50,
'workers': 8}
Loading encoders weights from irse50!
Loading decoder weights from pretrained!
Traceback (most recent call last):
File "/root/test/encoder4editing-main/encoder4editing-main/scripts/train.py", line 87, in
main()
File "/root/test/encoder4editing-main/encoder4editing-main/scripts/train.py", line 28, in main
coach = Coach(opts, previous_train_ckpt)
File "../training/coach.py", line 42, in init
self.lpips_loss = LPIPS(net_type=self.opts.lpips_type).to(self.device).eval()
File "../criteria/lpips/lpips.py", line 23, in init
self.net = get_network(net_type).to("cuda")
File "../criteria/lpips/networks.py", line 14, in get_network
return AlexNet()
File "../criteria/lpips/networks.py", line 81, in init
self.layers = models.alexnet(True).features
File "/root/miniconda3/lib/python3.8/site-packages/torchvision/models/alexnet.py", line 63, in alexnet
state_dict = load_state_dict_from_url(model_urls['alexnet'],
File "/root/miniconda3/lib/python3.8/site-packages/torch/hub.py", line 528, in load_state_dict_from_url
return torch.load(cached_file, map_location=map_location)
File "/root/miniconda3/lib/python3.8/site-packages/torch/serialization.py", line 593, in load
return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args)
File "/root/miniconda3/lib/python3.8/site-packages/torch/serialization.py", line 762, in _legacy_load
magic_number = pickle_module.load(f, **pickle_load_args)
_pickle.UnpicklingError: unpickling stack underflow

Process finished with exit code 1